this is still useful 9 years after, thanks sir :)

Goat

Can I similarly build a prefix tree of all prefixes for T and then use it to check substrings (since a substring is a suffix of a prefix)?

best explanation ever

Thanks for making this video! Feels so much more simple when you explain it

the guy in purple is a mogger h should be a model

I must be wrong but I feel that the storage of the connections between node and edge should also comsume some space? So the real space comsumptioin should be larger than 0.5*m^2, especially for those not well patterned strings?

This is pure gold. Great explanation Thank you !

what if we dont have t. I mean what should i do if first comparison is mismatch? in good suffix rule

Harris Brian Gonzalez Cynthia Moore Larry

If only my proffessor was as half as you, i wouldnt find my self here trying figure out this toppic,thank you!

I didn't get why you have issue to find the shortest string? 3:38 ABB has 2 equal options merge with BBB (2) or BBA (2). But you can create multiple branching, OR just skip node and move to another, like combine BBB with BBA, because it has (2), and AAAB with ABB, so you left with AAABB and BBBA, that's equal 7.

Love the video, but I feel as though you skimmed over the lexicographic sort very quickly without much explanation. @1:27 would be great if you had an intermediate step to explain the sort. Maybe understanding a lexicographic sort is a prerequisite to learning this, but if you are trying to start from nothing, it would be great to visualize that sort process! Otherwise amazing video, and love the reversal explanation! So interesting that many other videos do not explain the reverse given that the transformation is almost useless without understanding the reversal!

Thank you, sir! As a senior PhD candidate, I can also benefit from your detailed and vivid video lecture.

"We studied naive exact matching.." try "In ADS1: Naive exact matching, we studied naive exact matching" or "In the previous video. ADS1: Naive exact matching, we.."

Thanks a lot Ben! The series of lectures are very helpful to better understand some basic and classical algorithms in read alignment.

wow....thank you soo much..this concept was explained so well

What if we use ternary search tree to build suffix array

awesome

The best explaination for De Brujin😃

Binge-wathcing this playlist, gratitude from Brasil

You doing a great jobs! Thank you ^^^^

Man, how is someone expected to memorize and effectively recall all these at will in a FAANG interview?

The MEM stuff is a beautiful concept. I think it should be a Leetcode problem.

Excellent video, and you are an excellent teacher! Thank you.

Pretty clear and understandable video. thank you so much!

Thank you! This helped me a lot

Make more videos you really clear my concept 😊

Fun fact: This is the algorithm that makes grep extremely fast.

Hmm...maybe I don't get something here but...why are you using this visualization to explain the operations? With the previous video on wavelet trees, I was expecting an explanation with a visualization of a huffman shaped wavelet tree. I think this would communicate the connection between BWT, Huffman codes and wavelet trees better. Correct me if I'm wrong but aren't 'modern' implementations only storing the Huffman shaped wavelet tree of the bwt and nothing else, not even the C array?

Questions for the slide at 22:00 Let's say I take the string 'mississipimississippi' and use the codes on the slide to encode it. How would that look like? Do I just concatenate the codes from the codebooks accordingly? If yes, let's say I trasmit this encoded sequence of bits as well as the codebooks. Now the question is: If I only have the encoded string and the codebooks and I want to decode it, how would I do that? The codes inside the codebooks for each character are prefix free but of course they are not across all codes for all characters. When reading the bitsequence, how do I know from which codebook I have to chose the code from in order to decode, when I don't know which character I'm expecting or which character appeared before? I need to start somewhere. Let's say, I somehow know the first character 'm', I could then read the next (few) bits and then iterate over the codebooks to find out, if there is a character that has the 'm' as the left context and matches with the bit(s) I just read...but that feels a little inefficient... Also, what does '(no code)' mean in pratice, when encoding a string? I feel like I'm missing some context to fully understand what's going on here...

2:34 for anyone using this as a test case, the numbers are not how many places you should shift. They are only how many comparisons you are avoiding, which isn't as useful to know when coding. ( Algorithms class homework is to compare the pattern with the final characters, then work backwards to the beginning. I'm not going to give anyone code, but for a tip or two just relpy )

There's a small mistake: B.select(B.rank(i) - 1) is actually not "current or previous bit" but "previous bit" (<). B.select(B.rank(i)) in contrast is "current or next bit" (>=). If one wanted to achieve current or previous bit one could use B.select(B.rank(i) - (1 - B.access(i)))

well explained!

thank you. it was helpful

one of the best lection about CMS. However one question remains... how to calculate W and D if we know the element we will put inside the sketch (or total element of the stream)

The second video of his my engineering prof sourced us 😂

Do we use a terminator in each cycle because it wouldn't be possible for the DNA polymerase to know when to stop building the complimentary DNA? or is there another reason pls Thank you so much for this amazing work and for putting it out here for free.

It is the best course I had in the field of bioinformatics. Thanks for the great effort.

The material link doesn't work

Thank you for such a nice video. I have been improving my programming following the videos. I also tried using a different method to read the fastq file to extract the sequences and quality score. def readfastq (filename): with open (filename,'r') as f: file=f.readlines() seq=[file[i].strip(' ') for i in range(1,len(file),4)] qual=[file[i].strip(' ') for i in range (3, len(file),4)] return seq, qual

how powerful should be my machine to perform this tasks?

Fantastic material - thanks for making it public!

thanks for making wonderful vids, pls keep making more algorithms 👌❤

Awesome videos. Thanks a lot.

the way you visualization of hashing is so brilliant . thank you very much :)

Shouldn't (A,G) and (C,G) be swapped? EDIT: Just click the 'more' dropdown and you will see there's an amendment

Very clear, thanks very much!

oomg mark zuk

superb